Our People
Dr Simon Driscoll
PhD Student
Research interests
I am interested in using machine learning to learn sub-grid scale thermodynamical processes (those of melt ponds) in Arctic sea ice – this is important as 1) many scientific problems are not necessarily amenable to empirical models/derivations based on first principles, 2) melt ponds play a crucial role in the Arctic’s energy balance and 3) many climate models lack a melt pond parametrisation. Incorporating hybrid machine learning-data assimilation techniques, our work will create a new parametrisation of melt pond processes to be included in climate models around the world.
Recent publications
Verification of AI–based environmental forecasting systems: What can we do, what do we need to do, and what are the challenges?. 2026-06
DOI: https://doi.org/10.1016/j.jemets.2026.100032
Inductive Biases for Robust Climate Emulation Across Forecast Timescales. 2026-03-14
DOI: https://doi.org/10.5194/egusphere-egu26-16240
Textbook and code: AI for climate scientists. 2026-03-14
DOI: https://doi.org/10.5194/egusphere-egu26-20494
Weather and Climate: Applications of Machine Learning and Artificial Intelligence. 2026-03-14
DOI: https://doi.org/10.5194/egusphere-egu26-21143
Midlatitude Cyclone Intensity Biases in Machine Learning Weather Prediction Models. 2026-03-13
DOI: https://doi.org/10.5194/egusphere-egu26-3361
Observational data of Arctic Sea Ice Melt Ponds: a Systematic Review of Acquisition and Processing Approaches. 2025-10-09
DOI: https://doi.org/10.5194/egusphere-2025-4480
Replacing parametrisations of melt ponds on sea ice with machine learning emulators. 2025-01-20
DOI: https://doi.org/10.5194/egusphere-egu24-10749
Data-driven emulation of melt ponds on Arctic sea ice. 2024-10-25
DOI: https://doi.org/10.5194/egusphere-2024-2476
Parameter sensitivity analysis of a sea ice melt pond parametrisation and its emulation using neural networks. 2024-07
Contact details
University of Reading
07935314940
